Skip to content
Snippets Groups Projects

Draft: usb xhci additions by NXP

This brings in a few (small) usb changes for the dwc3/xhci host-controller side that NXP uses in their tree. By reading their commit messages this might help with #197 (closed)

I know this is a bit fuzzy. If this MR help, I'll try to identify which of these patches has real impact. Or some of these look useful enough that I'll ask NXP whether they want to try to mainline them or whether they mind if I would.

Merge request reports

Loading
Loading

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
  • saw the HC dying with these changes too

  • I wonder if they help with coming out of sleep. I'll try them on my kernel because with WWAN active the L5 often hangs in resume.

  • now with the "CAS" quirk applied to USB2 devices as well, see below, I have not seen the HC dying. But:

    I then see the modem silently stopping to work. No indication in the logs, but being offline. phosh also doesn't notice. Restarting modemmanager brings it back.

    not sure about whether that can be done.

    --- a/drivers/usb/host/xhci-hub.c
    +++ b/drivers/usb/host/xhci-hub.c
    @@ -1886,7 +1886,6 @@ int xhci_bus_resume(struct usb_hcd *hcd)
     
                    /* warm reset CAS limited ports stuck in polling/compliance */
                    if ((xhci->quirks & XHCI_MISSING_CAS) &&
    -                   (hcd->speed >= HCD_USB3) &&
                        xhci_port_missing_cas_quirk(ports[port_index])) {
                            xhci_dbg(xhci, "reset stuck port %d-%d\n",
                                     hcd->self.busnum, port_index + 1);
Please register or sign in to reply